A Bayesian Network Approach to Multi-feature Based Image Retrieval

  • Qianni Zhang
  • Ebroul Izquierdo
Conference paper

DOI: 10.1007/11930334_11

Part of the Lecture Notes in Computer Science book series (LNCS, volume 4306)
Cite this paper as:
Zhang Q., Izquierdo E. (2006) A Bayesian Network Approach to Multi-feature Based Image Retrieval. In: Avrithis Y., Kompatsiaris Y., Staab S., O’Connor N.E. (eds) Semantic Multimedia. SAMT 2006. Lecture Notes in Computer Science, vol 4306. Springer, Berlin, Heidelberg

Abstract

This paper aims at devising a Bayesian Network approach to object centered image retrieval employing non-monotonic inference rules and combining multiple low-level visual primitives as cue for retrieval. The idea is to model a global knowledge network by treating an entire image as a scenario. The overall process is divided into two stages: the initial retrieval stage which is concentrated on finding an optimal multi-feature space stage and doing a simple initial retrieval within this space; and the Bayesian inference stage which uses the initial retrieval information and seeks for a more precise second- retrieval.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Qianni Zhang
    • 1
  • Ebroul Izquierdo
    • 1
  1. 1.Department of Electronic Engineering, Queen MaryUniversity of LondonLondonU.K.

Personalised recommendations